Live Vinyl MP3

Contingency and Dispersion

Dropping the needle into narratives of circulation, the transmission-bound qualities of contingency, noise, and dispersion inform the theoretical inflections of this chapter and shape the delivery of its content. Over the course of what follows, I consider a set of files moving through two little databases that are likely to be more well-known (PennSound and UbuWeb) alongside two sites that are likely to be lesser-known to readers (Mutant Sounds and SpokenWeb). Following these collections, I examine a remixing tool entitled MUPS, developed by Canadian poet and programmer David (Jhave) Johnston, which introduces a new interface to a select set of files hosted by the PennSound archive. Like the stories of “The Defective Record” and “Africa(n),” this route is plotted to address the contingent effects enacted by the movement of a set of MP3 files, but will arrive at its destination only by way of conclusion.

Following this trajectory, I argue, a user might come to an understanding of the MP3 file by tracking files moving through a series of online collections and the meaningful transformations they facilitate. By the same token, these little databases may be understood through the same audio files that circulate through them. These circulatory effects are mutually constitutive. Along the way, the chapter samples a related compendium of adjacent digital objects via linked or referenced text, image, sound, and movie files. Just as someone might “read” a magazine exclusively for the pictures, I imagine someone might read this chapter exclusively for the downloads, which offer an archival ambience extending beyond the articulation of my argument. My aim is to chart the passage of a specific constellation of audio materials through several little databases as a staging ground for objects dispersed by online collections in general. In this way, the chapter is premised on the comparative principles of selection, navigation, description, and distribution. Corresponding practices of comparative media analysis or close listening may then be performed by the reader of these pages and the objects radiating out from them. 

To address the sites featured here, I follow the thread of a single poet’s output in an attempt to tease out a description of each online collection, how they mediate the recordings they host, and how we might begin to understand the transformative effects of dispersion patterns in the little database through the narrative of these files. I’ve elected to follow the trajectory of bissett’s files in particular and their dispersion across these databases for two primary reasons. First, because their distribution and republication across a host of digital platforms represent a dizzying orchestration of transmissions: echoing from Mutant Sounds, to PennSound, to SpokenWeb, to MUPS, and beyond. And second, because, as the editor who initially found these files at Mutant Sounds and ushered them into new contexts through their situation at PennSound, I’ve had the opportunity to track their staccato movements across an array of digital milieus. Throughout, the contextual effects of movement are the focus.

Indeed, bill bissett may seem an unlikely poet for the scholarly analysis of digital collections by proxy, given the relative paucity of academic writing on his generically variant and wildly experimental works, especially in sound performance. This choice will hopefully come into view in its telling: it is useful to follow the reverberations of an illustrative thread. Not a special thread, as a symptomatic reading or the hermeneutics of suspicion might have it, but any thread whatsoever. Raveling a path through the database is perhaps all that a narrative format like the monograph chapter might attempt: in other words, the essay can be understood as a kind of test for trajectories through these networks. Alongside the chapter, we might add archival downloads, edited compilations, and a related scroll of images, sources, or hyperlinks. If there is anything to be learned from the sites examined in The Little Database, it is that any text may also contain a collection. This chapter offers one such correlate to the little database in the echoes it follows.

While the bulk of this chapter follows a narrative trajectory of bill bissett files online, the appended interlude investigates a related constellation of audio files from the PennSound collection that have been transformed by their digitization. In this way, the chapter aims to execute both depth and surface levels of echolocation. Each sample in the interlude could warrant a similar excursus. Both modes of engagement are necessarily incomplete: the depth model of the chapter is a contingent exercise following one set of recordings through a series of transformative contextual processes; the surface model presents a wider array of changes, and cannot linger too long on any single effect. By pairing these operations, I contend that the two methods might be productively employed in concert. The first prepares speculative, latent narratives that the reader may discover hidden within the second—each contingency singular, every effect significant. This approach addresses the circulation of digital objects in general while also remaining attentive to the particularity of any individual file. In contrast to the site-wide metrics and sweeping preservation practices examined in the previous two chapters, this chapter moves toward the potential for a close reading of historical aesthetic artifacts as circulating digital objects in use, each with its own modulations to contextual and material significance. Taken into the interlude, these same files come to live in the everyday archives of user hard drives, played back in ordinary experience. While these personal collections remain beyond the scope of public access addressed by the collections tracked here, they all present similarly singular inflections to the files they host, opening on untested horizons of use and unknowable contextual registers as radically transformative as they are academically inarticulable.

As a supplementary gesture, the arguments in this chapter are sequentially interrupted by images of pages from books by bissett to encourage adjacent forms of reading. This formal exercise links the chapter to the image-based collection at Eclipse, while also leading into the network tracing of films hosted on UbuWeb in the following chapter. The images ground bissett’s audio recordings in a related material form. This relation of document to recording is, by turns, standardized and incommensurate. The text guides readerly performance, but the performance always exceeds the strictures of the text. In this regard, I suggest that the audio recordings are not connected only to vinyl records and digital files, but also to an expanded corpus of published and unpublished material texts and scripts by bissett. As an editorial premise, I have constrained these excerpts to works that relate to bissett’s poetics of audio performance and media recording in the late 1960s and early 1970s. Each image may be incorporated into the argument of this essay by the reader however they may choose to do so, even if I may have other suggestive connections in mind through their compilation. In league with bissett’s expansively permissive poetics, this chapter contends for a radically open text for qualitative listening and interpretive cruising. Following on what Mathieu Aubin has called “queer sonic resonances” in bissett’s literary recording, an audio recording accompanies each visual and paragraph, which may be heard in concert with both the featured images and the chapter.10 If the internally standardized queer orthography and performance practices in bissett’s output over the last thirty years have modulated the urgency of his poetics, perhaps these archival materials can be released anew—in facsimile and recording, altered but indexed—as prescient documents for an increasingly regulated media environment in the present, another way to cruise utopia. Taken together, I have found that this expanded set of investigations presents an argument for bissett as an important entry point into the study of digital transcoding and the queer temporality of afterlives generated by online dispersion.

Unexpected vectors and unimagined questions emerge when tracing the digital correlates to bissett’s profoundly experimental approach to analog media. However, a central query remains constant: what are the variable poetics that digital circulation introduces to a heterogeneous array of dispersion effects? My approach attempts to channel a mode of fidelity to the ways in which bissett’s early work, in the words of Darren Wershler, “defies conventional notions of genre: collages are paintings and drawings bleed into poems turn into scores for reading and chant and performance generates writing bound into books published sometimes or not.”11 As a kind of parasite on bissett’s web presence, the reader might consider this chapter to be the introduction to a new set of versions of these objects as digital files. The lines between preservation and republication are increasingly difficult to define. We might not consider a record transformed based on the library or archive that contains it. However, if the same album were republished by a new label with new liner notes, its bibliographic codes would be widely recognized as significantly revised. As is the case with the reissue of an out-of-print album, bissett’s output acquires new audiographic codes in each of the following sites of its reappearance. The little database, even masquerading as a chapter, also suggests that a highly significant mode of versioning is taking place in every instance.

Variable Outliers

Before launching into this contingent narrative, some brief remarks on alternate trajectories of the chapter are in order. From a digital humanities or cultural analytics perspective, each of the following databases seems primed for a range of analytic and metric tools for analysis. On the most basic level, a number of textual analyses seem readily available. For example, in sprawling collections like Mutant Sounds, UbuWeb, or PennSound, a simple set of maps and graphs that chart the geographic locations of recordings would be of interest. Through these data sets, we might query the geographic tendencies of these little databases. Where are most recordings made, why are specific locations best represented in these collections, and how might prominent outliers speak to these trends? The answers to these questions may indeed produce compelling results. Perhaps there is a bump in the number of recordings from Colorado given the Jack Kerouac School for Disembodied Poetics’ predilection for archiving. Or, we might imagine the role that New York City plays in the distribution of cultural capital even in allegedly horizontal online archives. More obviously, a great number of recordings are made in Philadelphia, given the documentary proclivities of the Kelly Writers House and the networks of affiliation with readings series in the city more broadly. A periodical reading might chart the exponential growth, stagnation, or decline of these collections, as seen in chapter 1. More pressingly, one might chart the representations of gender, race, or class throughout each collection, as the VIDA Count and others have done with literary magazines.12

The resulting metrics would serve the important function of confirming what a knowledgeable user might already suspect: Philadelphia is overrepresented in poetry readings; New York City maintains its prominence in recordings, even in its hosting of Bay Area poets; the distribution of gender, race, and class skews heavily toward white male readers of relative privilege; collections grow, stagnate, and ultimately cease operations. This chapter does not seek to confirm or dispute these matters of fact. There are systemic problems in these datasets that extend far beyond the metrics of digital humanities. If anything, the selection of bissett—famously and troublingly mischaracterized by Jack Kerouac as “an Indian boy; . . . Bill Bissett, or Bissonnette,” often disregarded by the poetic present, a queer polymath recorded variously as rock singer, concrete poet, or chanting artist—is highlighted for his variable outlier status.13

In another, more technical direction, a recent series of research projects has questioned how the digital humanities might begin to analyze digital audio collections of poetry. The most prominent of these is the HiPSTAS project spearheaded by Tanya Clement using ARLO software developed by David Tcheng.14 Rather than analyze metadata like location, date, or author embedded in sound collections, this project aims to work with the data of sound itself. The ARLO interface allows researchers to search and collate information that uses aggregate sound signatures as search queries. Kenneth Sherwood sums up the process succinctly: “At a simplified level, ARLO works by producing images of the audio spectra and then comparing these visualized time-slices with others across a range of pre-selected audio files.”15 Although ARLO was originally developed to aid ornithological research, poetry scholars have adopted the software to experiment with how we might investigate sound on its own terms. Interesting experiments have emerged: Chris Mustazza discerns the difference between aluminum records and magnetic tapes; Eric Rettberg has isolated patterns of laughter; Sherwood analyzes variant recordings of poems; Clement and Stephen McLaughlin visualize spectrograms of applause.16 Its possibilities are myriad. For example, AI-aided accuracy in speech-to-text recognition could open new pathways of use, and a more robust technical vocabulary for analyzing tenor and pitch could aid in the study of aural practices in poetry. While the potentials in this approach are promising, practical applications toward the advancement of critical scholarship seem a ways off, despite advances in audio machine-learning tools. This chapter is, once more, pushing against analytical aims. Rather than test new software for the processing of sound collections, I aim to chart the effects upon a specific set of files alongside current platforms for reconfiguring the little database as it stands. This is not to deny the promise of computational approaches to the collection as a whole. Instead, this chapter articulates a history of the circulation of sound in the recent past of the internet and the recomposition of the collection as a poetic project today, as told through the stories of the files it might host.

Mutant Sounds

The incident that catalyzed my interest in writing about file transmission through bissett’s work occurred fifteen years ago, when I stumbled on a relatively unnoticed archival release through a little database called Mutant Sounds.17 In the heyday of the “music blog” (once a prevalent mode of sharing sound files online), Mutant Sounds delivered an incredible array of rare and obscure albums, primarily ripped from out-of-print vinyl LPs, free for download. Founded in 2007, the Mutant Sounds site on Blogspot had amassed over three thousand posted releases in five years. Many were so rare that only the most devoted and well-resourced crate diggers might have ever heard their sounds otherwise. The reader might note that UbuWeb hosts just under one thousand entries in the “sound” subsection of the site and PennSound features approximately six hundred author entries. While these pages most often host multiple sets of recordings per entry, the volume is roughly comparable to the Mutant Sounds inventory. It’s all the more remarkable that an independent, noninstitutional initiative like Mutant Sounds, which ran on a free blog platform periodically releasing download links enabled by free file-sharing servers, might rival these collections as one of the great digital repositories (or, more boldly, archival collections) for obscure experimental recordings from the last half century. At least, it may have been recognized as such by its devoted core of users. However, these qualifications have become moot: the collection disappeared from the internet just as suddenly as it had once appeared.

In the spring of 2013, the website had recently ceased operations. After six years of reissuing out-of-print albums on a (misinformed) “notice-and-takedown” principle of online copyright law, a greater current led by the RIAA (Recording Industry Association of America) and the scandalous failure of Megaupload brought about the site’s demise.18 RapidShare, the “file locker” of choice for Mutant Sounds, had deleted most files that the site had uploaded to its server in an effort to avoid the same fate as Kim Dotcom.19 This process was accelerated by the emergence of a range of streaming platforms, from Pandora to Spotify, as well as a few musicians asserting a wide range of copyright claims over time. Currently, the Mutant Sounds catalog of thousands of digitized recordings lies in the obscurity of digital ruin. The site proves an important point that Vint Cerf, among others, has made over the years: an impending digital dark age threatens all online production; it’s best to print important works to analog media.20 We might add two additional points: one, that this type of collection is increasingly unlikely to emerge; and two, that the DIY internet of the 1990s and 2000s has long since disappeared.

On a roundtable hosted by The Awl and aptly titled “The Rise and Fall of the Obscure Music Download Blog,” Mutant Sounds editor Eric Lumbleau addresses the unique moment of these “sharity” sites, active primarily from 2004 until 2012.21 Summarizing the purpose of the Mutant Sounds collection for the Free Music Archive, the editors characterize their project as a campaign for “enlightening the masses to elusive musical esoterica buried beneath canned historical narratives and induced cultural amnesia.”22 The ideological thrust of the collection, much like UbuWeb and PennSound, was to destabilize the historical narrative by distributing an alternate canon far and wide. In particular, Mutant Sounds was plotted to combat the accepted progression from rock to punk to post punk, and to diversify these heterogeneous global forms. A brief scroll through their posts quickly presents a much stranger and far more diverse sense of experimental music from the mid-1960s to the present.

Or rather, one might hear that argument if the collection were still intact. Instead, today’s user will encounter only the contexts, descriptions, and images of albums one might never find elsewhere. Of course, the position of an album as charted by Mutant Sounds produces a dramatically different register from the presses with which—and poets with whom—even bissett’s experimental recorded work is most often associated. Nevertheless, with bibliographic attention to release dates, edition numbers, and the technical process of the digitization, the text still manages to transmit the essential components of archival metadata, also swiftly recoverable with a Discogs search. Perhaps these remnants of the site may still function like the Nurse With Wound list that guided the collection itself: as a series of signposts for further exploration beyond the regulated audio streams of Spotify and iTunes.23

Within this catalogue, on February 1, 2008, Mutant Sounds released a digital version of the singular album Awake in th Red Desert, recorded by bill bissett and th mandan massacre in 1968. At the time, I ran a pseudo-anonymous sharity blog of my own in between audio engineering sessions for PennSound, where I had recently become an editor. I downloaded the album immediately, as part of a habitual acquisition session among the various music blogs I followed. Before the “filter bubble” of social media encapsulated the navigable internet, these sessions were a mode of discovery within an enigmatic and unpredictable network. Like those done by many of the untold numbers of collectors tracking the releases shared by Mutant Sounds, my private little database grew within the contours of my own interests.

Unlike the stable collections of PennSound, UbuWeb, or SpokenWeb, these releases came to exist only as a dispersed set of objects on hard drives accumulating heterogeneous materials in unknowable configurations. In my own collection, Awake in th Red Desert arrived as a RAR archive file (Roshal ARchive; for Eugene Roshal, who developed the format in 1993). The contents of the decompressed folder were saved to my general “Music Library” folder and the RAR file was discarded. Duplicate copies of the album were placed in two adjacent folders: the first for potential upload to PennSound; the second in a database structured for the collaborative Endless Nameless project I published with James Hoff at the time. Endless Nameless organized digital objects by their original publishers to produce a series of limited edition hard drives. In this instance, I made a new folder entitled “See/Hear Records.” I rechart this activity, as it is quickly disappearing, if not entirely foreign, in an era of streaming sound and authorized in-application cloud-based purchases.24 The contingent provenance of other iterations, within other users’ collections, is as variable as the unknown numbers of downloaders.

Aside from the adjacent domain of experimental music, one primary difference from the collections that follow is that Mutant Sounds exclusively presented download links to full albums for external use. I’ll return to this point as it pertains to the other sites discussed, but for now, we might sketch the general character of this distribution method. Each blog post was linked to a full album download. These downloads were typically delivered as RAR or ZIP files that compressed a folder containing individual tracks, along with extremely lo-fi images of album artwork. The archive was a conduit for personal use, with the emphasis exclusively placed on sound. In the case of Awake in th Red Desert, this meant the exclusion of the “recorded book” that was originally distributed with the album by See/Hear Records. In this way, the fundamental purpose of the original publication, relating the printed page to the audio recording, had been excised from the digital release. This action is at the heart of the formlessness of the MP3. Jonathan Sterne has noted: “At the psychoacoustic level as well as the industrial level, the MP3 is designed for promiscuity.”25 Put differently, the sustained attention of “reading along” is counterintuitive to a format built for distracted listening and streamlined distribution. Recordings of poetry readings in the MP3, as we’ll see, typically present an alternative to these popular (distracted) uses of music online. Nevertheless, after download, the MP3 files were at the user’s disposal in their own private collection, ready to be played on any platform in any number of circumstances.

PennSound

When I first listened to the files, as the reader of this chapter is advised to do, I quickly recognized the singular contribution that Awake in th Red Desert might be heard to have made within the archive of countercultural poetics in the 1960s. I moved quickly to get selections from the album up on PennSound, where they might be distributed to a very different community of listeners. As a first step, I wrote to bissett directly for permission. In an important note on copyright, PennSound, unlike Mutant Sounds, operates as a strictly permissions-based platform. Bissett responded with characteristic charm and inimitable orthography: “yes xcellent if yu want 2 put seleksyuns from awake in th red desert on PennSound that wud b awesum[.]”26 To augment selections from Awake in th Red Desert, I decided to “segment” a full-length reading bissett gave at the Bowery Poetry Club in 2006.27 Both recordings were hosted with embedded links to individual MP3 files. For some time, this was the extent of PennSound’s bill bissett collection. Since then, the page has accumulated (among others): a movie recording of the 2012 Book*hug launch of its republication of Rush: What Fuckan Theory; a radio program recording from around 1978 through the Robert Creeley collection; an interview with Phillis Webb from the CBC in 1967; and, most recently, an untitled home recording from May of 2022.28 On another PennSound page, the user may also find a bissett track from the Carnivocal: Celebration of Sound Poetry album from 2004. From the Bowery Poetry Club back to See/Hear Records, the situation of Awake in th Red Desert on PennSound reveals myriad differences from the original upload.

While Mutant Sounds and PennSound hosted precisely the same bissett MP3 files—aside from the altered ID3 tags—the textual conditions of these two iterations could not be more different.29 In “Making Audio Visible: The Lessons of Visual Language for the Textualization of Sound,” experimental poet Charles Bernstein maintains that “the sound file exists not as a pure acoustic or sound event—an oral or performative event outside textuality—but as a textual condition, mediated by its visual marking, its bibliographic codes, and the tagging we give to it to mark what we consider of semantic significance.”30 An additional aspect of this condition, to recap previous invocations of Jerome McGann, is the social text. In other words, a more nebulous array of factors including reception, circulation, and context also inflect an object’s textual condition. To repost is to transform: not superficially, but at the most basic levels of a work’s significance. Here, we might note the immediate transformation from the lyrical “spew” in the music-based collection at Mutant Sounds to the institutional stamp of innovative poetry on PennSound. Where PennSound focuses on individual MP3s of poems and complete readings, Mutant Sounds released only full album collections previously published by a broad range of labels. Both sites push against canonical formations in their own way: Mutant Sounds against the tidy lineages of popular music; PennSound against the tidy lineages of lyric poetry. A confluence of these generic interventions inheres in the work of bissett. Awake in th Red Desert works in both directions, and each site transforms the historical recording in its own way, for its own listeners. If, as Steve McCaffery has argued in Sound Poetry: A Catalogue, bissett’s pathbreaking approach to sound poetry with Th Mandan Massacre was “significant in pushing poetic composition into the communal domain,” we might wonder in what domain it exists today.31

Despite their contrasting contexts, these two sound-based sites share a great deal in common. Like Mutant Sounds, the PennSound collection periodically releases audio recordings via a highly compressed MP3 audio format that typically remediates original cassettes, reel-to-reel tapes, vinyl records, and radio broadcasts. Both collections grow incrementally with each new digitized release, from a full series to an incidental recording. Primarily delivering files compressed to just 128 kilobytes per second, the MP3 actually prevents PennSound from receiving official recognition as a poetry recording archive.32 Where other projects might acquire certain types of funding for digital archives hosting “lossless” formats like WAV or FLAC (free lossless audio codec), PennSound’s emphasis on speedy distribution via MP3 files precludes it from archival classifications, despite the rarity, range, and depth of its collection. Instead, like Mutant Sounds, the PennSound collection is built for user downloads.

This approach reflects the uncertain future of the internet when the collection was founded by Bernstein and Filreis at the University of Pennsylvania in 2003. Even then, the MP3 offered a better alternative to the proprietary Real Audio format.33 Given its portability and accessibility, the MP3 remains the most popular format for audio distribution online. Widely circulated arguments on the distraction inherent to the MP3 are weakened by the “close listening” techniques that poetry generically warrants. The recordings at PennSound earn this attention in contexts as diverse as private academic research, university classrooms, and MOOC discussion boards. In remarks on the use of its thousands of poetry recordings, Filreis notes that the files have been downloaded by hundreds of millions of users. More promiscuously, we might imagine a million new versions across blog posts, syllabi, remixes, and other uses of the files.34

UbuWeb

Although Awake in th Red Desert is not hosted on UbuWeb, there are two illuminating mentions of the album among the twelve hits for the string “bill bissett” on the site; both citations are from discographies. The first, compiled by Michael Gibb and originally included in the remarkable Sound Poetry: A Catalogue, is titled “Sound Poetry: A Historical Discography.” The second, compiled by Dan Lander and Micah Lexier for Sound By Artists, is titled “A Discography of Recorded Work by Artists.” Of course, bissett is both poet and artist. However, the difference between the two—paired with the generic equivalence played out across the site—clearly bespeaks an entirely new condition for hearing bissett’s album. UbuWeb offers a less focused conjuncture: neither the musical archeology of Mutant Sounds nor the poetry-reading compendium of PennSound. Here the work joins with a wide range of objects beyond classification: conceptual art, structural film, concrete poetry, and other works that might be read under the sign of the avant-garde, broadly construed. Other hits for “bissett” include articles by McCaffery and derek beaulieu, a few concrete poems featured in various collections, bpNichol’s homage sound poem “Bill Bissett’s Lullaby” from Motherlove (1968), and bissett’s track “The Mountain Lake.” The last of these was recorded with guitar, tape, and “flux” as a contribution to an audio supplement to the 1984 sound poetry issue of The Capilano Review.35 Gathering this scattered assemblage of hits for bissett across the various sections of the site delivers a collection marked by the same intermedial disregard that characterizes the internet at large.

However, the most interesting audio work by bissett on UbuWeb does not appear in any of the search results. Clicking through the extensive sound section of the site, a user might stumble upon the entry for Past Eroticism: Canadian Sound Poetry in the 1960s (1986). The page hosting this digitized cassette simply breaks the audio into two MP3 files (Side A and Side B) in a technically convenient and tellingly remediated fashion. Aside from the title, there is no searchable text on the page. Instead, a JPG scan of the liner notes displays: “bill bissett (recorded 9/28/66* & 6/26/67**) / 7. Air To The Bells/The Face In The Moon** / 8. Valley Dancers* / 9. Colours*.”36 These three tracks immediately predate Awake in th Red Desert and anticipate that album’s generic freedom at the intersection of poetry, chant, song, and music. All three have been segmented for the first time for this page. Compared to the relatively metadata-scarce PennSound, the UbuWeb distribution of the MP3 file is even further stripped of context. And yet, an expansive interdisciplinary reading of the bissett files emerges through UbuWeb. If the files themselves carry little information, the context of their dispersion supports a robust network for an array of avant-garde productions in poetry, dance, sound, film, essays, radio, posters, and so on. All of this is compiled and released together with the happenstance bricolage of an assemblage magazine. Here, the bissett tracks merge with an undifferentiated conflux of historical and contemporary practices, dislocated from the original social texts of the work. By betraying all formal contexts, UbuWeb creates a digital context aptly suited to the genre-blending work bissett set out to record.

SpokenWeb

It is in relation to these three sites that I first considered bissett’s performance in 1969 at Sir George Williams University (SGWU), as presented on the SpokenWeb digital poetry archive.37 Recorded just one year after the release of Awake in th Red Desert, the SGWU reading draws from the same body of work that bissett deploys in his previous “recorded book.” As SpokenWeb tells us, with superb bibliographic attention, this reading by bissett features poems to be published in Nobody Owns th Earth (Toronto: House of Anasi, 1971), the lost angel mining co. (Vancouver: Blew Ointment, 1969), and OF TH LAND DIVINE SERVICE (Toronto: Weed/Flower, 1968). This last title is the organizing principle of the SGWU reading. It is also the only book released by bissett in the same year as Awake in th Red Desert. While the tape cuts off the opening, we can safely assume bissett begins with the first two poems in OF TH LAND DIVINE SERVICE, given the pattern of works to follow: namely, poems “3,” “4,” and a variation of “5” in the cycle, followed by intermittent works from across the book and concluded with “moss song,” the final poem from OF TH LAND DIVINE SERVICE. However, both the beginning and the end of the reading are cut off. The failure of the archive performs the play of in media res that bissett’s work invites. Sterne extends this aspect of the audio collection in his article “The Preservation Paradox in Digital Audio,” insisting that “sound recording is an extension of ephemerality, not its undoing.”38 Through no fault of the SpokenWeb editors, the lapse in the original mobile reel-to-reel tape recording renders the attentive metadata approach to this particular reading necessarily incomplete. There is no introduction to the reading; nor is there any commentary from the audience that we might analyze.39 The technical flaws in the reel-to-reel recording match with the aesthetics of assemblage, trans-genre performance, and the intermedial nature of bissett’s poetics. In other words, to borrow a track from Wershler, “bissett’s experiments on poetic excess yield highly specific social, historical and technological information about the shape and boundaries of what constitutes the permissible” in the contemporary audio archive.40 In this way, finally arriving at SpokenWeb, bissett delivers a compelling case for practices that operate at the limits of the little database.

It’s useful to query these limits of the database with regard to SpokenWeb, if only because the collection is itself so attentive to its materials. In its initial conception, it was designed to “develop coordinated and collaborative approaches to literary historical study, digital development, and critical and pedagogical engagement with diverse collections of literary sound recordings from across Canada and beyond.”41 SpokenWeb “begins with the preservation and description of sonic artifacts that have captured literary events of the past, and quickly moves into a wide range of approaches and activities that activate these artifacts in the present.”42 These scholarly approaches coded into the interface present an apparatus for critical engagement with the materials they present that is deeper than the minimal contexts supplied by sprawling MP3 repositories like PennSound or Mutant Sounds that are built for access above all. This depth model is characterized by the site’s reflexivity, including sections that feature research perspectives, audio analysis and sound visualization resources, ongoing events and blog posts, oral literary histories, and commentary on other audio collections online. Further, all of these features are concentrated on a set of recordings from a single reading series held at SGWU between 1965 and 1974. Each reading is in turn broken down with introductions, bibliographies, transcripts, sources, and an array of metadata on the recording. Annie Murray and Jared Wiercinski have extensively charted the development of these features of the site, and their papers on the subject are essential reading for the future of audio studies on the internet.43 This chapter, inspired by the expansive scholarly approaches of SpokenWeb, has been prepared to relate the traversal of a single recording adjacent to the SpokenWeb collection in order to sound out the relation of audio files hosted by Mutant Sounds, PennSound, UbuWeb and beyond into the private little database of any potential listener.

The differences between these sites and the SpokenWeb platform are, of course, quite pronounced. However, the commonalities they share may prove to be just as illuminating. SpokenWeb, like Mutant Sounds, sets out to map a network of relations in a given era of poetic production. Relating a community of practitioners in proximity to SGWU to an international group of poets and interlocutors, SpokenWeb presents the reading series as an “enormously intertextual affair,” following on Robert Scholes and Clifford Wulfman’s work discussed in the previous chapter.44 The series is a sounding board for developing poetics and unlikely combinations. In this way, bissett is linked to his contemporaries at the height of his investigation into the technologies of print publication and public performance. Like PennSound, the little database is distinguished by its focus on the poetry series, a periodical collection not unlike the complete run of a little magazine. Rooted in a specific locality with a concrete set of recording devices and live contexts, hosting the poetry series online amplifies an approach to the event of literary versioning. In Bernstein’s words, considering the sound file as part of the work “disrupt[s] even the most expansive conception of versions, all based on different print versions.”45 Both sites work to make this disruption possible. SpokenWeb, like UbuWeb, recodes its materials within an expanded set of concerns: bissett’s reading, like others featured on the site, is suddenly absorbed into an editorial argument on the scholarly use of digital platforms, the audio collection as an object of academic study, and the emerging challenges of studying sound with digital analytics. Surely, this is the most unlikely context for these readings. Who might have imagined in 1969, when the recordings were made, that such an inquiry would be the framework through which these readings would be received by a public audience?

Listening to the poetry series at SpokenWeb today, it’s impossible to ignore the depth of the original run, with its unique constellation of readers and the wider poetics community at SGWU. Of course, the poetry reading constructs its public. But it is equally impossible to erase the broader digital milieux and cultures of use in which this collection’s digitization surfaces. Thus, to listen deeply to a reading that bill bissett gave in 1969, the user must also hear a range of contemporaneous works hosted across the internet, to consider the digitization of the reading on a synchronic plane that includes versions circulating in unknown locations and unknowable configurations.

For example, further afield the user encounters samples of bissett’s “an ode to d a levy” from Awake in th Red Desert looping through The Chemical Brothers’ electronic music record We Are the Night. More abstractly, one might turn into an invocation of “Awake in th Red Desert” by Vancouver art rock band Dada Plan on the album DANCE MIRAAJ. Or, coming back to this page, this chapter, in the time since an early article draft of this text appeared online, the complete ambient image dataset running alongside these paragraphs has been deployed by VisPo artist Jim Andrews to populate an interactive bissett visual poem engine.46 Samples, evocations, remixes, to mention only a few examples of expanded use, all write and rewrite these same files into platforms, formats, and genres that can be written back into the narrative above only after the fact. Their inflections on the story lie latent within the ongoing flux of dispersion. The ordinary hard drives of an unknowable numbers of listeners enact their own idiosyncratic inflections to these files as they play through earbuds or reside within desktop folders. Vernacular practices of making, recording, posting, and playing online extend this story to a conclusion that can be only an open ellipsis . . .

Beyond bissett, these same playful encounters with transmission and use could be recounted through any number of recordings on SpokenWeb, UbuWeb, PennSound, and Mutant Sounds, among others. For this reason, rather than bind any of these threads into a single string, this chapter aims to trace, select, collect, edit, and disperse. Against a pointed argument concerning the ways in which bissett might illuminate the audio–visual–textual confluence on these sites, this essay points to a variety of alternate readings that bissett might enable the user to consider. This, I argue, is at the core of the digitized poetry reading. Already a kind of offshoot or supplement to the printed work, the social text of the historical reading radiates out to a wide range of materials connecting the past to our present moment, on- and offline, transforming our understanding of each in turn. The core of SpokenWeb’s design seems to be premised on this argument. As a platform for scholarship, the site directs its user outward: to read historical publications, critical articles, technical details, and source materials into each recording; to read the culture of the online audio collection into the poetry series; to read the potential for a future use of digital tools into a little database of audio recordings; to read the digital file, the poem, the book, and the reading at once. In each, a contingent set of concerns, highlights, and transformations arises by attending to the conditions of any given file. This scholarship is built on the sheer potentiality of reading the narratives generated within the endless versioning processes of the internet. As such, like the audio recording itself, it remains in the realm of the virtual: the pleasure of knowledge is joined with the impossibility of any full realization.

MUPS

In one notable afterlife to the transmission narrative above, the bissett files I’ve traced in this chapter could be found streaming within an interface called MashUPS (MUPS), written by David Jhave Johnston, released in 2012, and rendered inoperable in 2021.47 MUPS was an interactive flash platform for the live remixing of 1,260 audio files culled from the PennSound collection.48 Jhave describes the project as developed both “for the sheer pleasure of simultaneity” and “as a digital augmentation in the study of prosody,” where sites like PennSound “permit innovative explorations into the evolution of poetics.”49 As both transformative artwork and substantial act of digital scholarship, the project bears a resemblance to the textwarez works produced by Sebastian Lütgert. In one elegantly minimal page, the interface completely reorients our perception of PennSound. No longer is the site a repository for isolated audio files. Instead, it is an interface that affords the potential for a scenario wherein “poems speak to each other and with each other” through a widely variable set of parameters for computational audition.50 Each of the previously described bissett recordings from both the Awake in th Red Desert LP and the 2006 Segue Bowery Poetry Club reading are present on MUPS. If there were a fitting conclusion to the story of these bissett MP3s originating at Mutant Sounds, it would surely be the frenetic chorus generated by MUPS.

Taking this ending as canon, I conclude this chapter’s exploration of bissett-based transmission with MUPS. I offer this reading of the interface as a generative direction for thinking through the simultaneity of creative making and digital humanities approaches to the study of poetry recordings on the internet. In one of the few reviews of this work, Leonardo Flores similarly contends that these kinds of tools “aren’t just literary expressions informed by each writer’s poetics, they are also poetically and artistically motivated computational tools for some kinds of analysis associated with digital humanities methods.”51 As a mode of digital humanities making, Jhave’s MUPS performs an operation on PennSound that is both difficult to pinpoint as scholarship and relatively unrecognized as a work of poetics. It also harbors and remixes a little database of its own design. Both of these frames, I contend, are essential to understanding MUPS and its relation to PennSound. As additive layers, both of these frames also bear on the bissett files that have been included in the interface.

Before returning to bissett, we can start by outlining the technical details of MUPS. On loading the page, the user sees a 28×45 grid of black squares beneath a large display of “A———————————> Z” written in light grey, with a “mups” just above the Z. The page features a full-screen flash video program entitled “pennsoundup_WEAVE.swf” and a short Google Analytics tracking script. Hovering over any single square reveals author and file information, parsed and organized according to the naming conventions of the PennSound collection.52 Clicking on any square immediately begins the playback of the linked sound file, which is lit up in shades of red. A circle also appears to the left of the grid, which animates a visualization of the soundwave around a vertical volume slider. Clicking on subsequent squares plays corresponding sound files simultaneously. Concurrently played files stack up to the left of the grid as a complete sequential record of all recordings activated during a given session with MUPS. When a file completes playback or is clicked off, it turns to grey. Throughout, only when two or more files are playing simultaneously, the user will notice a text reading “WEAVE is OFF” beneath the volume slider to the left. Clicking on this text activates WEAVE, the most fascinating feature of the interface.

Turning on WEAVE activates an automatic switching mechanism between simultaneous files, based on perceived silences and intervals within audio recordings. The parameters of these switches are set by the user in three categories adjusted by vertical sliders: “threshold,” “tolerance,” and “pause.” First, “threshold” delimits the decibel level under which a sound recording is perceived as “silent” enough to switch. Second, “tolerance” sets the number of threshold points before a switch occurs. And third, “pause” sets the amount of time before the program looks for a new switch in each sample. In each parameter, the interface encourages quick shifting with notes like “HINT: To make sounds shift quicker, put TOLERANCE down.” Indeed, quicker skips are more impressive, as the user listens to the software jumping rapidly from one reader to the next with what can seem like the seamlessness of natural conversation, concerted collaboration, or an attentive DJ mixing the tracks.

With WEAVE activated, the user can hear “up to 32 streams” of poetry recordings in a simultaneous reading that shifts from file to file according to the parameters set by the user. Altogether this interface elegantly delivers the defining features of “new media” according to Lev Manovich: numerical representation (highlighting numerical values for switches); modularity (every sample can be resampled); automation (the program runs forward without input); variability (each listening is newly forged based on the inputted parameter); and, ultimately, transcoding (gathering sound artifacts from previous media performance formats). Indeed, Manovich’s own Soft Cinema is a near analogue to the generative listening system presented by MUPS. These works might be considered a form of database poetics in their reflections of the formal properties of the databases and algorithms that both determine and facilitate digital use.

However, in addition to the database poetics of MUPS, Jhave also executes a highly delimited editorial function. To facilitate a navigable user experience, MUPS distills the tens of thousands of sound recordings on PennSound down to 1,260 selected files. The selection process coheres only within the affordances of the interface. The “pleasure of simultaneity” is crafted from a diversity of voices, genres, reading styles, and recording textures. Historical recordings of Guillaume Apollinaire and Amiri Baraka interrupt contemporary readings by Yusef Komunyakaa and Nicole Brossard. Smoothly intoned science fiction narratives by Samuel Delany intersperse with the glitch aesthetics of Morris described at the opening of this chapter. The staccato punch of Christian Bök’s sound poetry might burst into ambient poetics by Tan Lin. Non-native English speakers (Rosmarie Waldrop), distinctive styles (Bernstein), and sound poetry tracks (Steve McCaffery), in particular, are heavily represented. More to the point, musical outliers to the PennSound collection are overrepresented in the selection: sizable segments of the grid are taken by radio artist Gregory Whitehead, the punk poetry album Redoing Childhood by Kathy Acker, and the audio-experimental-theatre production of San Francisco’s Burning by Helen Adam, each of which include musical accompaniment to the spoken word. These tracks are not the most representative of PennSound, but rather present samples from the limits of the collection. The editorial premise seems to argue that layering musical and sonically adventurous work produces the most interesting results for simultaneous mashups, or perhaps these works serve as analogues to the MUPS project as a whole. Within this setup, it is not difficult to imagine why bissett might command so many squares. The howl of “2 Awake in the Red Desert” or the moog synthesizers of “Now According to Paragraph C” provides the perfect counterpoint to more normalized patterns in the performance of poetry. In this way, MUPS facilitates the listening pleasure of formal discordance.

The little database deployed by MUPS differs from every other audio collection detailed in this chapter. The collection has no pretension toward completion. It offers neither the collector’s audiotopia of Mutant Sounds nor the comprehensive poetry catalog of PennSound. It ranges even further from the catch-all avant-garde of UbuWeb or the single-series depth of SpokenWeb. Like an extended mixtape, MUPS sacrifices the exhaustive in favor of the extraordinary. The interface works only if every track selected by the user is capable of generating compelling results within its internal network. There are scant traces of bibliographic or contextual data, such as the occasional listing of a place or series written into the PennSound filename itself. There are no links to works beyond the self-enclosed poetic system of MUPS. More than a collection or a poetic work, Flores deploys Judy Malloy’s useful summary of “authoring systems” in electronic literature to consider how MUPS is both software produced for user authoring and an “authored” work of software and selection in its own right.53 Using Malloy’s framework, Flores notes that Jhave “could’ve easily used this engine to create an e-poem or a series of them: expressions of the tool and his vision. Instead, he released the tool for users to have their own creative explorations and analysis of the material.”54 However, these explorations and analyses are circumscribed within a highly curated set of audio samples and, most importantly, a tightly coordinated set of interface options for playback. Like the rigid typographic structures of bissett’s printed texts, MUPS presents a stable interface for unpredictable performances.

Where Mutant Sounds, PennSound, and UbuWeb facilitate file downloads, SpokenWeb and MUPS afford specific modes of playback alone. In both projects, the website is the primary source of audition. Mutant Sounds is located on the opposite end of the spectrum, exclusively offering options for download. PennSound and UbuWeb fall somewhere in between, with immediate listening as easily accessible as the capability to download. On SpokenWeb, the user is presented with a highly compressed visualization of audio waveforms as a static image within the play-bar (SoundCloud is the popular analog to this interface, which is immediately recognizable as a tool for navigating a sound file on the internet). MUPS, on the other hand, visualizes the waveforms of any given file only in the moment of its transmission. Like iTunes visualizations of sound files or the animations once common to CD players, the representation on MUPS is purely aesthetic. Unlike the rare files offered exclusively on SpokenWeb, the MUPS interface operates parasitically on PennSound. With an extra step, the user can use PennSound to find background information on any recording, as well as a direct link to download the MP3 file.

By contrast, PennSound operates on the dual principles of accessibility and depth. All files are freely accessible, for both course syllabi and private browsing. Increasingly, many files are linked to close readings and expanded materials via the collaborative scholarship of PoemTalk; the interview format on Close Listening; the breadth of critical writing on Jacket2; the learning environment presented by the ModPo MOOC; the textual materials offered at EPC and Eclipse; or any number of related online class syllabi.55 While MUPS is a stand-alone interface, it feeds into the wider realm of little databases linked up to PennSound’s pedagogical model of distribution.

In certain respects, this extension loses sight of the purpose of MUPS. While the interface may be a conduit for students or scholars to discover new works of poetry or delve into a deep array of academic resources, this function is not primary to the project. Instead, MUPS emphasizes the possibility for improvisation within a given script. In this regard, it is exactly aligned with the live performances of bissett, wherein typographically fixed concrete poems become fluid chants and intonations. Utilizing the strict protocol of written script for loose improvisation, the interface returns the free potential for transformation to the fixed MP3 recordings of bissett. Put differently, the structural framework of MUPS serves as a corollary to bisset’s improvisatory aesthetics. Two files from Awake in th Red Desert might jump back and forth in microsecond intervals. A temporal jump from 2006 back to 1968 could happen as easily as a jump between contemporaneous recordings by bissett and his collaborator bpNichol. MUPS offers all the technique of a skilled DJ simultaneously managing thousands of vinyl recordings for the casual user with the single click of a cursor. While media critics often use the example of sampling techniques or remix aesthetics to describe database culture, the acts of spoken improvisation and reading performances are less often invoked as an allegory. However, this is precisely what bissett on MUPS performs: neither the static representations of the page (or digital image) nor the rigid grooves of vinyl (or MP3), but rather the whimsical sampling of audition and attention in the context of a sound-based social text.

Within the contextual and procedural system generated by MUPS, bissett’s live performances meet an interface scripted for the corresponding live performance of its listener. The “poetry reading” on playback through MUPS directs the user into the aleatory poetics of reading as a contingent practice. Afforded by the “promiscuity” of the MP3 and built on the modular variability of its digital processing, MUPS returns bissett to the site of performance and demands that the listener develop new protocols of improvisatory listening. As a work of creative scholarship, this project plays the little database as a reading practice rooted in the same free jazz that backs a number of tracks on Awake in th Red Desert. As scholarship, it asks—and answers—questions that the essay format could not begin to address. From the outset, the authorial figure of bissett is but one potential mode of clustering a set of files circulating online through these platforms. This interface could be read differently through the poetics of any other grouping of tracks in the collection. How does MUPS address the ambient stylistics of Lin or the palindromic utterances of Whitehead? What about the five recordings of “Ursonate” on MUPS, each with a different interpretation of the score? How might all sixteen tracks recorded in 1968 speak to the interface? Or to the sixty-nine tracks recorded in 2008, for that matter? The MUPS interface begs us to ask these questions, and remains open to the intervention of any given user. Even as we’ve long contemplated the relation of narrative to database, these signifying systems can be difficult to visualize, and even more difficult to do so succinctly, playfully, and concretely. Compelling systems for simultaneous forms of playback, rendered in a random-access format, remain rare to come by for practical use. In MUPS, a close listener might wind up creating a “live” MP3, playing echoes among the little databases simultaneously.